Off-Label Use of Pharmaceuticals: Trends and Drivers1
W. David Bradford∗
Department of Public Administration and Policy, and
Department of Economics, University of Georgia
John L. Turner†
Department of Economics, University of Georgia
Jonathan W. Williams‡
Department of Economics, University of Georgia
April 2014
Abstract
We identify the rate of off-label use of prescription drugs in the United States during 1993-2008. We apply Detection Controlled Estimation to a comprehensive crosssection of prescriptions and find that rates of off-label use rise from 30.2% to 39.1%
during this period. This coincides with a surge in settlements of Department of Justice
lawsuits for off-label marketing. Additionally, physicians are more likely to prescribe
off-label when there are fewer FDA-approved alternatives and when a patient’s insurance has less restrictive formularies and lower copayments. These substitution
patterns are consistent with off-label prescribing by physicians that enhances the welfare of patients.
JEL Code: L51, I10, I18, I13, K23
Keywords: off-label use, pharmaceuticals, detection controlled estimation.
1
We thank Guy David, Paul Greenburg, John Rizzo, Matthew Rosenberg, Geoffrey Sanzenbacher, Tamar
Sisitsky, Marta Wosinska, and participants at the UGA/GSU/Emory Health Economics Conference for
helpful comments. Remaining errors are our own.
∗
School of Public and International Affairs, Baldwin Hall, Athens, GA 30602. Phone: 706-542-2731.
E-mail: [email protected].
†
Terry College of Business, Brooks Hall, Athens, GA 30602. Phone: 706-542-3682. E-mail: [email protected].
‡
Terry College of Business, Brooks Hall, Athens, GA 30602. Phone: 706-542-3689. E-mail: [email protected].
1
Introduction
Since 1962, the US Food and Drug Administration (FDA) has restricted the marketing of
a drug to just the set of “on label” indications for which the drug is approved. However,
physicians may prescribe any approved drug for any indication.2 In the market for pharmaceuticals, which accounted for $321.3 billion in sales in the United States in 2010 (2.2%
of US GDP),3 “off label” use is common and potentially desirable. On one hand, the best
treatment for a patient’s particular indication may require using a drug off-label. In addition, applying FDA-approved drugs to new uses may be a particularly cost-effective type of
innovation because these drugs have already passed safety benchmarks in clinical trials. On
the other hand, ineffective off-label use is socially wasteful. In a few cases, off-label use has
also led to patients being physically harmed.
Not surprisingly, off-label use is enormously controversial in the clinical and policy communities and among federal regulators (Salbu 1999; Klein and Tabarrok 2004; Stafford
2008). And yet, no prior research supports broad, systematic, and trend-based analysis of
off-label use. Indeed, no economics papers analyze it empirically, and virtually nothing is
known about its welfare consequences. In this paper, we start to fill these gaps.
We apply Detection Controlled Estimation (Feinstein 1990) to a comprehensive data
set of patient prescriptions from the National Ambulatory Medical Care Survey (NAMCS)
during 1993-2008 to identify the incidence of off-label use and to test for what drives it.
We find that in the most recent years, more than one in three prescriptions are written
for off-label uses. This rate is lowest in the earliest years, and rises from 30.2% in 1993 to
39.1% in 2008, a 29.6% increase. The three years with the highest frequency are 2006, 2007,
and 2008. Coincidentally, this is also a period in which off-label use has been part of the
national policy conversation due to settlements between pharmaceutical manufacturers and
the U.S. Department of Justice (DOJ) on cases involving off-label use claims (Kesselheim,
2
For clarity, we use the term “indication” instead of “diagnosis,” “condition,” “disease,” etc., throughout
the paper.
3
Source: 2010 IMS National Sales Perspectives.
2
Mello and Studdert 2011).4
Perhaps most importantly, our estimates show patterns of off-label use that are consistent with choices we expect rational, fully-informed patients to make. For example, after
controlling for a host of drug and patient characteristics, we show that a ten percent increase in the number of drugs that have been approved to treat a patient’s set of diagnoses
leads to a 4.7% reduction in the probability that a physician prescribes off-label. We also
find that when patients face lower out-of-pocket costs, physicians tend to prescribe off-label
more often. Relative to patients with no insurance, patients with insurance (of any type) are
more likely to be prescribed off-label. Of those with insurance, those with Medicaid are the
most likely to be prescribed off-label. Specifically, relative to those with private insurance,
the probability of a physician prescribing off-label is about 2.2% higher when a patient is insured through Medicaid. This may be in response to the relatively weak prior authorization
programs at the state level for Medicaid since 1990 (Dranove, 1989; Huskamp, 2003) or the
very low copayments (relative to privately insured patients) that Medicaid recipients pay.
Those on Medicare are prescribed off-label at a similar rate to those on private insurance,
which is consistent with the privatized nature of “Medigap” policies that were in place over
much of our sample time period (Oliver, Lee and Lipton, 2004; Rowland, 2001). Together,
these substitution patterns are consistent with the hypothesis that off-label prescribing by
physicians enhances patient welfare. To our knowledge, this is the first research to shed
light on how off-label use affects welfare.
Studying off-label use retrospectively using prescription data presents two significant
challenges. First, we must classify uses as on-label and off-label. Unfortunately, no existing
archive tracks a drug’s FDA-approved uses across time. We use annual issues of the Physician’s Desk Reference (PDR) to build yearly matches between drugs and their approved
(on-label) indications. To match the non-standardized indications in the PDR to the list
of International Classification of Disease - 9th Revision (ICD-9) codes from NAMCS pre4
Combining cases from our own search with those from Kesselheim et al. (2011), we identify 33 DOJ
settlements during 2004-12.
3
scription records, we rely on a professional Clinical Documentation Specialist employed at
a major academic medical center hospital. We treat a drug as being on-label for an indication if it has the same active ingredient as one of the drugs identified in the PDR as being
on-label for that indication.
Second, the way that NAMCS prescription data (and data in nearly every other survey or
retrospective data set) are recorded almost guarantee false detection of off-label use. For one
thing, indications recorded on survey forms are limited by the number of available fields. For
another, physicians often base their reports on administrative claims on which indications
will maximize reimbursement.5 Perhaps most importantly, patients frequently visit their
physicians about one indication and receive a prescription for another. For example, suppose
a person with chronic hypertension visits the physician because he has the flu. If the
physician does not record hypertension as an indication on the NAMCS form, but does
record a (convenience-driven) re-fill prescription for an ACE inhibitor, then naı̈ve inspection
of the prescription record would classify the use of the ACE inhibitor as off-label. This issue
is problematic in nearly every other clinical, administrative or retrospective data source.
To overcome these problems, we appeal to Detection Controlled Estimation (DCE), first
used by Feinstein (1989; 1990; 1991) to study tax evasion and regulation of U.S. nuclear
power plants. Intuitively, this procedure constructs a model that separately predicts the
probability of on-label use and the probability of whether it is detected, with estimation
done via maximum likelihood. Crucially, identification requires a subset of variables that
affect only the probability of on-label use and a subset of variables that affect only the
probability of detection. For the latter category, we rely on the plausible assumption that
changes to the NAMCS survey form affect detection but do not affect a physician’s decision
to prescribe on-label. The form permits up to five prescriptions during 1993-94, up to six
from 1995-2002 and up to eight from 2003-2008. Since the maximum number of prescriptions
5
In a study providing advice to physicians on how to increase revenue through careful attention to
reimbursement intricacies, Heidelbaugh et al. (2008) states “Be sure to list active and acute medical
indications discussed during the visit... rather than those that are stable...”. Again, this could cause a
prescription to appear off-label in claims data, even if it were actually prescribed for an approved use.
4
and indications is often binding, and an increase in the number of prescriptions relative to
the number of indications tends to increase the likelihood that on-label use is not detected,
these changes vary the rate of detection exogenously.
More work is required to fully understand the efficiency and welfare properties of off-label
prescribing. For example, our paper does not identify either the dollar costs or treatment
gains of off-label uses relative to on-label alternatives. We view our results, which identify
the overall rate with which the practice occurs and identify some determinants of physicians’
decisions to prescribe off-label, as an important first step in understanding the phenomenon
of off-label prescribing.
The remainder of the paper is as follows. Section 2 provides some background on the
practice of off-label use in the pharmaceutical industry, including a literature review. We
discuss our data and model in Sections 3 and 4, respectively. Section 5 discusses our results
and Section 6 concludes.
2
FDA Oversight and the Nature of Off-Label Use
The Food and Drug Administration Act of 1906 created the FDA, initially just to set
manufacturing standards. In the wake of the Elixir Sulfanilamide episode, Congress passed
the Federal Food, Drug and Cosmetic Act of 1938.6 This prevents a new drug’s introduction
without FDA certification that the drug is safe. The law also led to the modern arrangement
where many drugs are available only with a physician’s prescription (Temin 1979).7
In 1962, Congress passed the Kefauver-Harris Amendments to the Food, Drug and Cosmetic Act. This grants the FDA the authority to certify a drug’s efficacy, in addition to
safety, before a firm can sell it. Initially, supporters of this law argued the law would prevent
6
In the 1930s, the S.E. Massengill Company sold the antibiotic sulfanilamide first as tablets and capsules.
They then developed a liquid version by dissolving sulfanilamide in diethylene glycol. In September 1937,
Massengill marketed this liquid as “Elixir Sulfanilamide.” Unbeknownst to their chemists, diethylene glycol
is toxic. Over 100 people died. See Temin (1979).
7
As Temin (1979) discusses, the new law did not directly implement the prescription system, but the
FDA moved quickly to use the law to establish this system.
5
firms from aggressively marketing drug products, with dubious effectiveness, to physicians
who would then write prescriptions for their patients. However, the Thalidomide episode
of 1961-62 created a sense of urgency that also helped facilitate passage of the law (Harris,
1964). While the 1938 law kept Thalidomide off the US market, some physicians had already received the drug for experimental purposes. The FDA did not heavily regulate this
type of distribution. Reports of birth defects in babies born to European mothers who had
taken Thalidomide raised concerns that pharmaceutical firms might harm patients by moving their products to market too quickly (Peltzman 1973). This series of events contributed
significantly to the emergence of the current regulatory environment, which has important
implications for the incentives for firms to gain approval for particular indications.
2.1
Drug Development
Drug development begins with the isolation of a new molecule. A researcher then tests
to determine whether the molecule is biologically active and to identify the nature of that
action. Once action is determined, usually in animal models, the researcher (typically backed
by a pharmaceutical manufacturer) files an investigational new drug application (IND) to
begin human trials. These clinical trials follow a strict three-phase process in which the
applicant must prove safety and efficacy.8 If successful in these trials, the applicant submits
a New Drug Application (NDA) to the FDA; if it is approved, the NDA allows the applicant
to sell the drug in the U.S. and specifies a list of approved indications. Firms may market
their product for the approved (i.e., on-label) indications only.
The process from IND to NDA is long, risky and costly. Typically, it takes a decade or
more (DiMasi et al., 2003), and only around 9% of drugs for which an IND is filed achieve an
NDA (DiMasi 2001). Because firms typically seek molecule patents at the moment of initial
discovery, the process often consumes half or more of the life of the core patent covering the
8
In Phase I, healthy humans are given the compound to establish safety. Phase II tests involve a small
number of volunteers with the disease for which the manufacturer is ultimately seeking approval. Phase II
aims to establish some measurable treatment effect. Once a minimal efficacy standard is met, the study
proceeds to Phase III, which requires much larger pools of volunteers with the disease be tested to identify
the treatment effect more precisely. All three phases are generally double blind.
6
molecule. Given failure rates, direct costs of drug discovery and testing, and the opportunity
cost of capital that must be devoted to the effort, estimated average costs of achieving an
NDA range from $800 million to $1 billion (DiMasi et al. 2003; Adams and Brantner 2006).
Because of the incentives in the clinical-trials process, firms do not typically include a
broad list of approved indications on their drugs’ labels. Part of the reason is that clinical
trials center on demonstrating efficacy for a given indication, and the results of one approval
(NDA) are often not transferable to another. To add a new indication for an existing drug,
a manufacturer must go through the same tedious, costly clinical-trials process needed to
achieve the original marketing approval.9 In addition, physicians may prescribe a drug
off-label anyway, in which case obtaining a new indication is unnecessary.
The FDA grants a variety of marketing exclusivities, some of which affect strongly the
incentives to pursue NDAs and new indications. First, a drug approved with a new molecule
may earn a new chemical entity (NCE) exclusivity. For five years, the FDA prevents any
other firm from marketing the drug, regardless of patent protection.
More importantly, the FDA also grants a three-year marketing exclusivity for a label
change that recognizes a new indication (NI). In principle, the NI exclusivity grants a
significant incentive to endure clinical trials to modify a drug’s label. However, this benefit
depends crucially on generic competition. If one or more generic manufacturers have gained
approval to sell a drug, then the benefits of obtaining an NI may spill over to generic
manufacturers. Intuitively, a firm that obtains an NI receives only a three-year exclusive
right to advertise its product for the new use. But since a physician can prescribe a drug
for any reason, if one version of a drug is advertised for a use, physicians may prescribe all
other versions at nearly the same rate.
Prior to 1984, generic competition was less of a concern. Because generic applicants had
9
As the USGAO has described it: “If, after FDA has approved a drug, evidence arises of its safety and
effectiveness in treating conditions or patient groups other than those named in the label, then the drug’s
manufacturer (or any other interested party) can submit a new application to have the label changed. This
application, known as an ’efficacy supplement,’ is similar to the original application in that it must contain
evidence demonstrating to FDA’s satisfaction that the product is both safe and effective for the treatment
of the new condition” (USGAO, 1996: page 2).
7
to go through clinical trials to sell drugs bioequivalent to approved drugs, and because there
was tremendous uncertainty about whether testing a drug infringed on patents, generic
products were a relatively small part of the approved drug portfolio in the US. In 1984,
however, Congress passed the Hatch-Waxman Act to clarify the rules for, and increase the
rate of, generic entry.10 Now, a generic applicant may receive an Abbreviated New Drug
Application (ANDA), permitting entry, by demonstrating that its product is bioequivalent
to the branded product. The cost of an ANDA is a small fraction of that of an NDA.
Generic entry has expanded so rapidly, that generics now account for nearly two-thirds of
all prescriptions written in the U.S. (Aitken et al. 2009).
Practically speaking, obtaining a new indication for a drug (and getting the three-year
NI marketing exclusivity) has high economic value when the drug is covered by an NCE,
moderate economic value after the NCE expires but patents still partially block generic
entry,11 and low economic value upon generic entry. Since some new uses emerge late
during a drug’s life cycle, manufacturers may often find it optimal not to pursue an NI
exclusivity.
As an alternative, a manufacturer may pursue approval for a new drug. For example,
Glaxo-Wellcome gained FDA approval for buproprion hydrochloride in 1985, called it Wellbutrin (IR formulation) and marketed it as an antidepressant.12 Some time after, Dr. Linda
Ferry, a family practitioner in California, noticed that many of her patients taking Wellbutrin (for depression) showed decreased interest in smoking. She then convinced Glaxo
to pursue clinical trials. After successful trials, Glaxo got approval in 1997 for an on-label
indication for buproprion as a treatment for smoking cessation. However, Glaxo did not
add smoking cessation to the label for Wellbutrin (and they never have). Rather, they got
10
Innovating firms can also earn (up to five years) patent life lost during the clinical-trials process.
Under the Hatch-Waxman Act, a generic may file for an ANDA once four of the five NCE years have
expired. It must then show the pioneering firm’s patents are invalid or would not be infringed by the generic
product.
12
Initially, there were a significant number of seizures at the recommended dosage of 400-600 mg. Glaxo
removed Wellbutrin from the market in 1986. After subsequently discovering that reducing the dose by
about half sharply reduced the risk of seizures, Glaxo reintroduced Wellbutrin to the market in 1989 with
a maximum dose of 450 mg/day. See http://www.emedexpert.com/facts/bupropion-facts.shtml.
11
8
smoking cessation on the label for a new drug, Zyban (Perkins et al. 2008, pp. 113-114).
Obfuscating the relationship between Wellbutrin and Zyban may mitigate the advertising
spillover effect.
2.2
Literature Review
Given the FDA’s regulatory architecture, there are a variety of reasons that drugs’ labels do
not cover all uses. Numerous papers discuss anecdotes of off-label use. Collectively, these
anecdotes illustrate that the desirability of off-label use varies widely.
For example, beta-blockers, such as metaprolol and propranolol, have been used for
decades to treat hypertension, cardiac dysrhythmias, and other diseases. Clinicians have
noted that beta-blockers also control physical sensations associated with anxiety (such as
rapid heartbeat, tightness in the chest, and trembling), and that when patients do not
feel these sensations, their psychological experience of anxiety is significantly reduced. As
a result, these drugs are widely prescribed for situational and other forms of anxiety, to
apparent great effect. Lin et al. (2006) estimate that 52% of prescriptions for beta-blockers
were off-label from 1999-2002. Anxiety is not, however, an approved use of any betablocker. Such examples of effective re-purposing are not rare and off-label use is not a new
phenomenon. In a 1991 survey, roughly one-third of cancer drugs were prescribed off-label
and about one-half of patients were prescribed at least one drug off-label (USGAO 1991).13
On the other hand, off-label use has led to clear cases of patients being harmed. During the 1990s, after Dr. Michael Weintraub showed that a group of 121 patients using a
combination of the weight-loss drugs fenfluramine and phentermine lost an average of 30
pounds, this “Fen-Phen” combination surged in popularity. Because neither drug’s label
discussed using the drugs in combination, this was an off-label use. Dr. Weintraub looked
for side effects, but he assumed the drugs were safe (Kolata 1997). Unfortunately, numerous
patients suffered heart-valve damage (O’Reilly and Dalal 2003).
13
Salbu (1999) and Klein and Tabarrok (2004) discuss a number of other examples of beneficial off-label
use.
9
There are also many cases, especially among the drugs involved in the DOJ settlements, where pharmaceutical firms are alleged to have marketed their drugs for off-label uses
with uncertain clinical support (Stafford 2008). In perhaps the best-known case, WarnerLambert’s drug Neurontin was initially approved for “adjunctive therapy in the treatment of
partial seizures...in patients above the age of 12 years,” and was later approved for patients
3-12 years old (2000) and for “postherpetic neuralgia” (2004). Physicians have prescribed
Neurontin for a number of off-label indications, such as bipolar disorder and neuropathic
pain, for which the evidence for effect is at best equivocal (Mack, 2003). In 2004, Pfizer
(which merged with Warner-Lambert in 2000) pled guilty to two felony counts of marketing
a drug for unapproved uses, and paid $430 million in civil and criminal fines. Some $26.6
million went to whistleblower David Franklin, who started working for Warner-Lambert in
1996 (Evans, 2009).
Numerous clinical papers study rates of off-label prescribing. Nearly all are narrowly
focused, and their results vary widely and are virtually impossible to compare to each other.
For example, in a survey of papers studying off-label use in pediatrics, Cuzzolin et al. (2003)
identify 16 studies published during 1995-2001. Of these, 13 are prospective studies, 2 are
retrospective, and 1 is prescription-event monitoring (a much longer study, 10 years). The
number of patients varies from 40 to 24,337, while the number of prescriptions varies from
257 to 4,455.14 The percentage of off-label use varies from a low of 10.8% (McIntyre et
al. 2000, “pediatric ambulatory”) to a high of 72% (Avenel et al. 2000, NICU). In the
most comprehensive work, Radley et al. (2006) study off-label use during 2001 of 160
drugs. Studying data from the National Disease and Therapeutic Index (NDTI), a survey
of physicians from IMS,15 they find off-label use at about 21% of overall use. This is near
the 32.5% rate we estimate for 2001, even though we examine a more comprehensive set
of drugs and use different (non-proprietary) data on prescriptions. Finally, USGAO (1991)
and Molitor (2012) examine some drivers of off-label drug use by cancer patients. USGAO
14
Some studies did not indicate numbers for both categories.
These data are “nationally representative diagnostic and treatment data similar to that contained in
the National Ambulatory Medical Care Surveys (NAMCS)” (Radley et al. 2006)
15
10
(1991) shows, for example, no pattern of off-label use by age group and gender, while Molitor
(2012) finds that over 20% of new cancer drug use within the Medicare population during
1998-2008 was off-label.
3
Data
Our data are drawn from two sources. The National Ambulatory Medical Care Survey
(NAMCS) provides information on the prescriptions written for a representative set of US
physician office visits. NAMCS data are collected for 1993-2008. The Supplement to the
Physicians’ Desk Reference (PDR) is titled either the PDR Companion Guide or PDR
Guide to Drug Interactions, Side Effects, Indications, Contraindications, depending on the
year. This serves as physicians’ primary reference on FDA-approved indications for each
drug, for 1993-2008 as well.
No existing research combines these two sources.16 Moreover, because of the incentives
facing firms to get indications on drug labels, the usually unanticipated timing of discoveries
of useful ways to repurpose drugs, and the ways physicians’ manuals (like the PDR) list sets
of approved indications, there are multiple ways of classifying on-label and off-label use. To
clarify, we first define two types of on-label use:
• A drug-label use occurs whenever a drug is used for an indication on the drug’s
FDA-approved label.
• An active-ingredient-label use occurs whenever a drug is used for an indication
on any label of any drug with the same active ingredient.
For any drug, the set of drug-label uses is a subset of the set of active-ingredient-label uses.
Next, we define two types of off-label use.
• An anticipated off-label use is any active-ingredient-label use that is not a druglabel use.
16
Radley et al. (2006) use the PDR to classify off-label use.
11
• An unanticipated off-label use is any use that is not an active ingredient use.
Legally, active-ingredient uses that are not drug-label uses are off-label uses. However,
these off-label uses are natural candidates for prescribing physicians given that the use has
been approved for some other drug with the same active ingredient. Hence, we say such
off-label uses (e.g., Wellbutrin used for smoking cessation) are anticipated. For the purpose
of our analysis, we define off-label use as unanticipated off-label use. All other uses of a drug
will be defined as on-label.
3.1
NAMCS
The NAMCS is an annual survey, conducted by the Center for Disease Control and Prevention (CDC). The multi-state design generates a representative sample of patient visits to community-based (non-institutional, non-clinic) physician offices. The data include
variables capturing office visits, indications, characteristics of patients, insurance coverage,
medical procedures, tests conducted, and medications prescribed. The survey has been
conducted annually from 1973 to 1981, in 1985, and then again annually since 1989. Each
year approximately 1,500 physicians participate in the survey. Each physician is randomly
assigned a one-week participation window and uses an abstraction form to characterize approximately 30 office visits. This results in detailed quantitative descriptions of around
35,000 office visits each year. Visit weights are assigned so that national estimates can be
produced.17
We extract data on all visits from 1993 to 2008. After we keep those visits with at least
one prescription, the resulting data set contains information on 547,977 prescriptions from
the 141,523 office visits during 1993 to 2008. For each visit, the physician may record up to
17
This use of the NAMCS weights to estimate prescribing volume has been validated in the literature;
see Pincus et al. (1998), Thomas et al. (2006), and Iizuka (2004). Note that the NAMCS visit weights are
calculated to impute prescriptions to the national and annual level. Given the weight construction we could
allocate annual prescription estimates to the month by dividing the weighted estimate by 12; allocating
prescriptions to the region could be accomplished by re-weighting on the basis of the region’s proportion of
the national population. In either case, any biases from not explicitly adjusting our prescription counts will
be isolated in the intercept term, and will not affect the marginal effects of interest.
12
Figure 1: The Current FDA Label for Creon (Pancrelipase)
3 indications. The NAMCS data code indications according to the ICD-9 classification. In
addition to prescriptions and indications, we extract information on visit payment source,
patient age and gender, physician specialty, and Census region.
3.2
PDR
Each prescription drug available for sale in the U.S. has an FDA-mandated label, which lists
(among many other things) the indications for which the FDA has approved use. Figure
1 is an extract from the first part of the Creon label. Its label shows the drug-label use
(treatment of exocrine pancreatic insufficiency due to cystic fibrosis, chronic pancreatitis,
pancreatectomy, or other conditions) in specific populations. Identifying the history of
FDA-approved indications for a drug is surprisingly difficult. Although the FDA maintains
the current FDA-approved indications for a drug in the Orange Book, these listings do not
include the history of how the label reached its current state.
The annual Supplement to the PDR fills this void. In particular, the “Indications” section lists FDA-approved drugs for each indication, giving physicians up-to-date information
on a drug’s approved uses, interactions, side effects, and contraindications. We obtain each
edition of the PDR from 1993 to 2008 and capture all drug-indication combinations.18 Fig18
To extract this information from over 1000 pages formatted in this way, we remove the Supplement ’s
binding and use a high resolution scanner to create images of each page. We use Able2Extract Professional,
software that uses Optical Character Recognition (OCR) technology, to extract the information to a raw text
format to parse into a useable form. Following the extraction and parsing processes, we search for and remove
13
Figure 2: An Example of an Indication in the PDR
ure 2 provides an example of the Supplement’s formatting from 1997. For the indication
“Pancreatic cystic fibrosis,” there are seven associated FDA-approved drugs: Cotazym Capsules, Creon, KuZyme HP Capsules, Pancrease Capsules, Pancrease MT Capsules, Viokase,
and Zymase Capsules. PDR records also include the active ingredient for each drug—in this
case, Pancrelipase for all drugs—and the manufacturers.
Note that this list represents anticipated off-label uses as being on-label. For example,
Creon is approved for “treatment of exocrine pancreatic insufficiency due to cystic fibrosis
or other conditions,” i.e., pancreatic cystic fibrosis. But Viokace is approved just for “treatment of exocrine pancreatic insufficiency in adults due to chronic pancreatitis or pancreatectomy,” i.e., not pancreatic cystic fibrosis. Use of Viokace for pancreatic cystic fibrosis, like
Wellbutrin for smoking cessation, is an active-ingredient use that is an anticipated off-label
use.
Clearly, anticipated off-label use is qualitatively distinct from unanticipated off-label use.
Often, manufacturers wish to market drugs with the same active ingredient differently, so a
drug’s labels do not include all safe and effective uses. As mentioned above, to most closely
the (surprisingly small) number of errors in the resulting files. The consistency in the Supplement ’s format
and few changes in the text descriptors for indications and drug names make this feasible. For example,
to catch errors in the text describing an indication, we sort the entire panel (all years) by indication and
flag subtle changes to the text descriptor. We then compare the flags to the original image files. We use a
similar cleaning process for drug names.
14
match the actual list of all safe and effective uses, we define any active-ingredient-label use
as on-label. Hence, we treat anticipated off-label use as on-label use. This provides a liberal
definition of on-label use and a conservative definition of off-label use.
We format the data so that the unit of observation is an indication-year-ingredient
(IYI) combination. For the text from Figure 2, the IYI is “pancreatic cystic fibrosis - 1997 pancrelipase.” There are 178,823 total IYIs, composed of 3,587 unique indication descriptors
and 3,342 unique active ingredients.
3.3
Matching NAMCS to PDR
Drugs and indications are each defined differently in the NAMCS and the PDR. Hence,
matching drugs to indications for the purposes of identifying off-label use requires careful
interpretation of language from both definitions. For drugs, we use a word-based matching
algorithm to match the text for each active ingredient in the PDR to the NCHS Multum
codes used by the NAMCS to identify active ingredients. This corrects for abbreviations
and is virtually one-to-one.19
Matching indications in the PDR to indications in the NAMCS is far more complicated.
The PDR uses terms for indications that do not map directly to the ICD-9 codes used
on NAMCS prescription records. In fact, processing the language in these descriptors and
matching them to ICD-9 codes is sufficiently complicated that we rely on a Clinical Documentation Specialist, or coder, from a large academic medical center.20 The coder manually
matches each PDR indication to the appropriate ICD-9 code(s).21
This process is often not straightforward. Some of the language in the PDR indications
requires further research for additional information to establish an appropriate match. In
other situations where the PDR indication is not specific enough to map to a single 5-digit
19
The few instances where the match is not one-to-one is due to combinations.
This person’s primary role at the medical center is to look for missing, vague, or incomplete physician
documentation and to query the physician for the specificity needed to ensure that the coders can capture
the most accurate indication, indication or symptoms for the patient.
21
Our coder was provided a spreadsheet with the 3,587 unique text descriptors from the PDR for 19932008, as well as the ICD-9-CM Manual Volumes 1, 2, 3 published by HCPro, Inc (2011).
20
15
ICD-9 code, we match to a less-specific (four- or three-digit) code corresponding to a broader
range of indications. For example, “Pneumonia, Community Acquired” and “Pneumonia
Nosocomial” are listed as separate indications in the PDR. “Pneumonia Nosocomial” is a
more complex pneumonia with an easily identified ICD-9 code for Gram Negative Pneumonia, 482.83. “Pneumonia, Community Acquired” is less specific and is coded as a Pneumonia, NOS (not otherwise specified), 486. In other situations, a descriptor requires more
than one ICD-9 code to adequately describe the indication. In these situations, the coder
lists all relevant codes.
Hence, matching PDR indications to NAMCS indications is a many-to-many match. Our
conservative matching approach also leads us to under count off-label use. Alternatively,
we could make subjective decisions regarding the closest matching indication. We regard a
conservative approach as most sensible here.
Having matched PDR indications to NAMCS indications and PDR active ingredients
to NAMCS active ingredients, we reformat our data so that the unit of observation is a
prescription. For each observation, we then list the characteristics of the patient, the date
of the visit and the complete list of indications assigned for the visit. For each prescription,
we cycle through the indications to determine whether the set of approved active ingredients
for those indications includes the prescribed drug. To be conservative in our estimates of offlabel use, we convert the NAMCS indications to the three-digit ICD-9 level before querying
the set of approved drugs. For example, 410.01 would be matched to any 410 code.
We set the on-label indicator to 1 if the prescribed drug’s active ingredient is among the
set of approved active ingredients for at least one of the indications listed for the visit. This
serves as the dependent variable in our analysis. As we discuss in more detail in section
4, this variable cannot be used to directly measure the amount of on-label use because of
imperfect detection.
16
3.4
Descriptive Statistics
Table 1 shows the demographic composition of the individuals and physician specialties
associated with prescriptions written in our sample from the NAMCS. Just over 58% of
prescriptions are written for female patients, 8.9% are for African-Americans, and 7.5% are
for Hispanics. The average prescription is written for a patient who is 51.8 years of age.
Over 93% of prescriptions are written for patients who have some form of insurance—47.5%
with private insurance, 31.3% with Medicare, 10.2% with Medicaid, and 4.1% with some
other form of insurance. The physician specialties with the greatest frequency are internal
medicine (13.2%) and cardiology (11.2%). The Midwest Census region accounts for the
largest proportion of prescriptions at 34.4%, while the East Census region (omitted region)
accounts for the lowest, only 20.3%.
Table 2 shows that the most commonly prescribed drugs are central nervous system
agents (20.2%), cardiovascular agents (18.5%), and anti-infective agents (10.5%). The most
common indications are associated with circulatory disorders (25.4%) and respiratory disorders (17.7%). The frequencies of the indications do not sum to one, because all indications
from a visit are listed for each prescription from the same visit.
On average, a patient is prescribed 3.9 prescriptions per visit. However, the standard
deviation is quite large. Across all visits, a physician has an average of 71.5 FDA-approved
alternatives for the entire set of (up to three) visit-specific indications, defined at the threedigit ICD-9 level, reported in NAMCS. The median number of approved prescription alternatives (i.e., unique Multum codes) per indication is 27.7. We expect that when the on-label
alternatives for treating a patient are limited (i.e., a small number of approved drugs for the
complete set of observed indications), a physician will be more likely to prescribe off-label.
The NAMCS survey form changes significantly over our sample. From 1995 to 2002
(40.1% of our sample), the form allows six prescriptions to be reported, up from only five in
1993 and 1994. From 2003 onward, the form allowed a total of eight prescriptions to be listed.
The form never increases the maximum allowable number of indications, three, during our
17
period of study. Over 61.7% of prescriptions appear on a form with two indications listed
and 33.5% appear on a form with three indications.
Table 3 highlights how the NAMCS form limitations affect detection. The numbers
in the table are conditional means of the on-label indicator, by numbers of prescriptions
and indications. If the form limitations have identifying power, then our ability to detect
on-label use should be highest for those visits with a low number of prescriptions and a
high number of indications (top right corner of the table), and lowest for those visits with
a high number of prescriptions but a low number of indications (lower left corner of the
table). These are precisely the patterns we observe. The conditional means of this indicator
rise virtually monotonically up the columns and across the rows, and plateau in the topright corner of Table 3. Thus there is a group of observations for which the NAMCS form
limitations appear to have significantly less impact on our ability to detect on-label use.
Hence, time-varying form limitations are very useful for identification purposes. Each
of the exogenous increases in the maximum number of prescriptions that can be recorded
was not accompanied by a similar increase in the maximum number of recorded indications
during a visit. These form limitations result in only 33.5% of all prescriptions in our data
appearing to be written for an FDA-approved use, or 66.5% off-label use.
22
In the next
section, we discuss how we correct for detection error and identify the rate of on-label use.
4
The Model
We cannot use our naı̈ve indicator of on-label use directly because of false negatives. When a
prescription is on-label but the indication is not listed on the NAMCS survey form, true onlabel use is unobserved. To identify the rate of on-label use, we adopt the DCE methodology
22
The most comprehensive of these studies, Radley, et al. (2006), estimates a rate of off-label use of 21%
for over 160 drugs in 2001; we estimate an off-label use rate of around 26.5% for that same year. However,
Radley and coauthors are only able to conduct the study for one year and had to rely upon an expensive
proprietary data set (the National Disease and Therapeutic Index (NDTI) from IMS Health) and a handgenerated indicator for on-label use. Consequently, their findings and methodology are both limited and
dated.
18
introduced by Feinstein (1990).23
Let a physician’s decision, of whether to write prescription i for a drug that is on-label,
be summarized by a stochastic latent variable,
Y1i∗ = x1i β1 + ε1i ,
(1)
where εi1 is independently and identically distributed N(0, 1). The binary outcome of this
decision-making process, which is not observed, is then
Y1i =



1 if Y1i∗ > 0
0 otherwise .
(2)
With the normality assumption, the probability that the physician prescribes on-label, conditional on x1i , is
P r(Y1i = 1) = Φ(x1i β1 ),
(3)
where Φ(x1i β1 ) is the standard normal cumulative distribution function (CDF). The variables in the linear index, x1i β1 , alter the likelihood of a physician writing a prescription for
an approved indication.
The difficulty is that we do not observe this binary outcome and cannot infer our outcome
of interest (i.e., Φ(x1i β1 )) by estimating a Probit model. Instead, we observe an indicator
that takes on a value of one when a prescription is both on-label and we can detect it in the
data. Conversely, the indicator takes on a value of zero when the prescription is actually
off-label or when we have a false negative. By specifying a model for detection, along with
the equation characterizing the physician’s choice to prescribe on-label, DCE allows us to
distinguish between these two cases.
Consider cases when on-label prescribing actually occurs and let
Y2i =
(
1 if on-label use is detected
0 otherwise .
23
(4)
DCE has been applied previously in a health context by Kleit et al. (2003) and Bradford et al. (2001),
and a non-health context by Helland (1998), Feinstein (1989), and Feinstein (1991).
19
Thus, when Y2i = 1, on-label use occurs (i.e., Y1i = 1) and is detected. This is precisely the
naı̈ve on-label indicator we construct. Further, assume the distribution of the underlying
latent variable for the Y2i indicator is given by
Y2i∗ = x2i β2 + ε2i .
We assume ε2i is distributed standard normal and the linear index, x2i β2 , includes those
variables that alter the probability of detecting off-label use.
The probability of detecting true on-label use is
P r(Y2i = 1) = P r(Y2i = 1|Y1i = 1) ∗ P (Y1i = 1) + P r(Y2i = 1|Y1i = 0) ∗ P (Y1i = 0).
Assuming there are no false positives, i.e., all detected on-label use is truly on-label, then
P r(Y2i = 1|Y1i = 0) = 0 and P r(Y2i = 1|Y1i = 1) = P r(Y2i = 1).24 With the normality
assumption,
P r(Y2i = 1) = Φ(x1i β1 )Φ(x2i β2 ).
The likelihood function is then
L=
Y
[Φ(x1i β1 )Φ(x2i β2 )]Y2i [1 − Φ(x1i β1 )Φ(x2i β2 )]1−Y2i .
i
As Feinstein (1990) points out, certain conditions must be satisfied to identify the parameters
of the model via maximum likelihood. In particular, note that when Φ(x1i β1 )Φ(x2i β2 ) is
high, it is unclear whether this is due to a high rate of on-label prescribing, or a high
rate of detection. For example, if x1i and x2i include all the same variables, then nothing
differentially shifts the probability of on-label prescribing and the probability of detection,
24
We assume that there are no false positives in our measure of on-label prescribing. Given the presence
of false negatives, this assumption is required for the detection model to be identified. In our application,
false positives can arise if there are errors in matching NAMCS indications to the PDR data. We hire a
professional medical records coder to do the matching to minimize this possibility. The other source of false
positives arises when a patient is actually prescribed a drug off-label for one observed indication but the
drug is approved for another of the patient’s observed indications. We believe the frequency of such events
is rare, and would lead to a conservative estimate of off-label use.
20
so β1 and β2 are not separately identified. But if there are some variables in the group
determining on-label prescribing but not in the group determining detection (and viceversa), then each probability will vary significantly conditional on a fixed value for the other
probability and the model is identified. Ideally, we would include a variable in x2i that
pushes the probability of detection to one, and identify the rate of on-label prescribing in
a simple discrete choice model. Without this variable, the parametric assumptions (i.e.,
normality and linear indices) are required for identification. In Section 5, we discuss how
close we get to this ideal setting.
The common elements in x1i and x2i move both the probability of on-label use and the
probability of detection. For this group of elements, we include patient characteristics such
as age, gender, and race. For example, a patient’s age may lead physicians to take a more
cautious or aggressive approach in prescribing off-label. Also, certain types of patients may
be more likely to have a chronic indication (e.g., African-Americans have a higher incidence
of diabetes), which decreases the probability of observing the relevant indication for the
visit, lowering the probability of detection. Those variables only in x1i move the probability
of on-label use but do not affect the probability of detection. These include the number of
FDA-approved drugs for the visit-specific indications, plus indicators for physician specialty,
categories of indications (e.g., mental disorders), the patient’s insurance status and census
region and year. Those variables only in x2i move the probability of detection but do not
affect the probability of prescribing off-label. These include the number of prescriptions
written on the visit, plus indicators for whether there were two or three indications and for
the maximum number of prescriptions recordable on the NAMCS survey form.
Thus our strategy for separately identifying β1 and β2 relies largely on exploiting the
limitations, and changes in those limitations, of the NAMCS form during our sample period,
along with some visit-specific information. Recalling Table 3, we expect an increase in the
maximum allowable number of prescriptions to decrease our chances of detecting on-label
use, but not to affect the physician’s actual decision to prescribe off-label.
21
With estimates of the β̂1 and β̂2 parameters, we recover the rate of off-label prescribing
as the complement of the rate of on-label prescribing, 1 − Φ(x1i β̂1 ). This can be calculated
at the individual prescription or population level. However, as noted by Feinstein (1990),
even when valid exclusion restrictions give consistent estimates of Φ(x1i β̂1 ), the coefficient
estimates on those variables common to x1i and x2i may not be precisely estimated, as identification relies largely on parametric assumptions. In general, we interpret these coefficients
with caution.
5
Results
It is simplest to discuss our estimates as two separate models, one describing detection and
the other physician behavior. Table 4 and 5 present the coefficient estimates for selected
variables in the model of detection and physician behavior, respectively. The top half
of each table contains the estimates for variables that are exclusive to each model, while
the bottom half of the respective tables presents estimates for variables common to both
models. Because identification of the coefficients for variables common to both models is
driven largely by the parametric structure of the model, we are hesitant to draw strong
conclusions about them. Tables 4 and 5 also have two columns. Column 1 presents the
results from a parsimonious specification, restricting temporal trends to be the same for all
physician specialties in the model of physician behavior. Column 2 presents estimates from
a more general model of physician behavior that allows for differential trends by including a
full set of interactions of year and physician specialty indicators. We emphasize the results
from Column 2.
It is useful to first consider the results for the detection model. The coefficient estimates
in Table 4 for the detection model are consistent with our expectations. Both the number
of prescriptions for a visit and the form changes to the NAMCS decrease the probability
of detecting on-label use, and each coefficient is statistically significant at the 1% level.
Further, the indicators for whether there are two or three indications for a visit are positive
22
and statistically significant, such that for a fixed number of prescriptions an increase in the
number of indications increases our rate of detection. The results in Table 4 are similar
in both columns. Collectively, the intuitive results in Table 5 give us confidence that we
successfully identify the rate of detection.
NEED TO MENTION THAT INCLUSION OF DUMMIES OR TIME TREND DOES
NOT CHANGE THE RESULTS, WHICH SHOWS THAT THE EXCLUSION RESTRICTION IS CORRECT AND WE’RE WELL IDENTIFIED.
NEED TO MENTION HOW THE PROBABILITY OF DETECTION CHANGES WITH
FORM CHANGES FROM COLUMN (2). Prob[detection — form1=0, form2=0] = .545104
Prob[detection — form1=1, form2=0] = .5169105 Prob[detection — form1=1, form2=1] =
.4895006
NEED TO MENTION HOW WE CAN INCLUDE PHYSICIAN SPECIALTY INDICATORS IN THE DETECTION MODEL AND NOTHING CHANGES FROM COLUMN
(3). ALSO MENTION THAT WE CAN ADD THE INSURANCE INDICATORS TO THE
DETECTION EQUATION AND ESTIMATES ARE STILL VIRTUALLY UNCHANGED
IN EITHER DETECTION OR ON-LABEL MODEL.
NEED TO SAY THE ESTIMATES ARE NEARLY IDENTICAL REGARDLESS OF
THE ROBUSTNESS CHECKS, SO WE FOCUS ON COLUMN (2).
Moving to the model of physician behavior, Table 5 reports how the rate of actual onlabel prescribing varies with the number of FDA-approved alternatives for the physician to
treat a patient. We find that an increase in available alternatives significantly increases the
probability that a physician prescribes on-label. The effect is also economically significant.
Increasing the observed number of on-label alternatives by 10% (about 7.2) decreases the
probability of prescribing off-label by 1.6 percentage points, from 34.0% to 32.4%. This
corresponds to a 4.7% reduction in the probability of off-label prescribing. This is, at a
minimum, consistent with welfare-enhancing behavior on the part of physicians. In going
beyond FDA-approved alternatives to find a good match between a patient and a drug,
23
physicians are more likely to prescribe off-label when their choices are more limited.
Table 5 also reports the estimates of the effect of insurance status on the physician’s
decision to prescribe on-label. Relative to the base case of a patient with no insurance
(the omitted dummy variable), patients with any type of insurance are significantly more
likely to be prescribed a drug off-label. We estimate marginal effects for insurance status
by assuming a particular insurance status for every patient, calculating the probability of
on-label prescribing, and comparing that to the probability of on-label prescribing when all
patients have a different insurance status. If no patient has insurance, our estimates predict
the proportion of prescriptions written off-label as 32.8%. If all patients have Medicaid, we
estimate the proportion as 34.8%. This corresponds to a 5.9% increase in the probability
that a prescription is written off-label for patients on Medicaid. In comparing patients with
Medicaid to those with private insurance, those on Medicaid are 2.2% more likely to be
prescribed a drug off-label. Results are similar when comparing to patients with Medicare.
These results are consistent with at least two behavioral hypotheses. First, physicians
may account for the insurance status of their patients when deciding whether to prescribe
a drug off-label (which may run counter to a patient’s insurance formulary or prior authorization program). Thus, consistent with the decision to go off-label when the portfolio of
approved choices is limited, the physician appears to have patients’ pecuniary welfare in
mind when writing prescriptions. Second, the difference in the effect of other forms of insurance and Medicaid suggests either that reimbursement for off-label prescriptions through
Medicaid is less constrained or that physicians respond to the very low (or no) patient copayments for prescription drugs in Medicaid compared to other insurance by more frequently
providing off-label prescriptions.25
Using the estimates of β1 and β2 from Column 2 of Tables 4 and 5, we construct implied
estimates of the probability of detection and off-label use. Table 6 gives the mean of the
25
Due to space considerations, we omit the remainder of coefficient estimates for the model of physician
behavior: indicators for physician specialty, indication codes, and census region and year. These estimates
are available from the authors upon request.
24
predicted probability of detection, Φ(x2i β2 ), for the taxonomy of prescription-indication
combinations from Table 3. Over all observations, we estimate that the naive indicator
is correct 50.6% of the time. The rate of detection falls significantly as the number of
prescriptions increases, given any fixed number of indications. For example, on average,
for visits with three indications and only one prescription, we detect on-label use 66.3%
of the time. Yet this probability falls to only 29.7% when the maximum number of eight
prescriptions is listed for a visit, along with three indications. This pattern holds across
all 3 columns of Table 6. These probabilities provide the average adjustment applied to
observations with certain combinations of number of prescriptions and indications.
NEED TO RE-ITERATE THAT WE GET A BIG BUMP FROM THE FORM INDICATORS ALONE HERE.
For some prescriptions, the detection rate is over 80%, which means that we have a
group of prescriptions detected to be on-label with a very high probability. The limiting
case where the probability of detection is one is accomplished through the model’s parametric
assumptions. The limiting case where the linear index equals infinity does not exist in realworld retrospective data, but observing very high detection rates for a subset of prescriptions
in our data gives us confidence in the DCE approach. A very cautious interpretation of
results would treat our estimates of off-label use as tight upper bounds on the true rate
since we get close to the limiting case in our data. Feinstein (1990) argues that this is how
DCE models should be interpreted.26
Figure 3 plots yearly rates of off-label use, along with a fitted polynomial. The rate
of off-label prescribing fluctuates between around 30% and 35% between 1993 and 2002,
and then rises consistently from 2003 to 2008. Overall, off-label use rises by nearly a third,
from 30.2% in 1993 to 39.1% in 2008. Interestingly, this detected trend in off-label use also
leads, by just a few years, the trend in off-label settlements, which accelerated from 2008
to 2011. Since our results cover essentially all drugs listed in the PDR, this gives us further
26
For example, in Feinstein’s (1991) study of tax evasion, it is necessary to assume that there is at least
one auditor in the data that detects evasion with probability 1 in the absence of parametric assumptions.
25
0.4
Mean Off-Label Use, 1 − Φ(x1 β1 )
0.39
0.38
0.37
0.36
0.35
0.34
0.33
0.32
0.31
1994 1996 1998 2000 2002 2004 2006 2008
Year
Figure 3: Trend in Off-label Use
confidence in our results over the 15-year period we study.
NEED TO PLOT THE ESTIMATES FROM EACH COLUMN TO SHOW THAT
THEY ARE VIRTUALLY IDENTICAL.
NEED TO SAY WE FOCUS ON ESTIMATES FROM COLUMN (2).
Figures 4(a) and 4(b) summarize the trends in off-label use by drug class; drug classes
with the highest percentage increase in off-label use are in Figure 4(a), while drug classes
with the lowest percentage increase in off-label use are in Figure 4(b). With the exception
of a few drug classes (coagulant, gastrointestinal, and psychological), each ends our sample
period with a higher rate of off-label use than it begins. Additionally, it is uniformly true
that those classes with the lowest rates of off-label use in 1993 had higher percentage increase
in off-label use. Thus the practice appears to be growing most quickly in drug classes where
it had been less common.27
27
Radley et al. (2006) is the only other study, to our knowledge, that makes any attempt to estimate
off-label use by drug class; however, we do not know the identify of the few drugs they study within each
class, making the estimates impossible to compare.
26
0.4
0.35
0.7
Anti-Infectives
Cardiovascular
Hormonal
Metabolic
Respiratory
Topical
Mean Off-labe Use, 1 − Φ(x1 β1 )
Mean Off-labe Use, 1 − Φ(x1 β1 )
0.45
0.3
0.25
Nervous System
Coagulant
Gastrointestinal
Immunologic
Psychological
Miscellaneous
0.6
0.5
0.4
0.3
0.2
0.2
1994 1996 1998 2000 2002 2004 2006 2008
Year
(a) Largest % Increase in Off-Label Use
1994
1996
1998
2000 2002
Year
2004
2006
(b) Smallest % Increase in Off-Label Use
Figure 4: Off-label Use by Drug Class
6
Conclusion
Off-label use is prevalent, controversial and under-studied. In this paper, we take the important step of identifying the incidence of off-label use from 1993 to 2008 and important
factors that drive it. We are optimistic that our approach and these results will lead to
further insights about this controversial subject and, hopefully, recommendations for policy.
For example, our finding that Medicaid patients are more likely to be prescribed offlabel is important in light of recent policy developments. As of 2008, Medicaid expenditures
on prescription drugs for certain low-income adults and children total $15.2 billion dollars
(USGAO, 2010: page 1); however, this program will be expanded significantly under the Patient Protection and Affordable Care Act of 2010. While our findings suggest that off-label
prescribing patterns are consistent with the enhancement of patient welfare, additional economic analysis is required to ensure that the practice is desirable from a societal perspective
and that tax dollars are spent efficiently. For instance, research that builds on our findings
might study whether treatment outcomes from using drugs off-label justify potential costs
(realized by any entity) in excess of on-label alternatives.
27
2008
In addition, the DOJ continues to enforce FDA guidelines that ban promotion of offlabel uses for drugs, but it does not know the extent to which this promotion induces
physicians to prescribe off-label or the effect that strict enforcement has on patient welfare.
Our finding that physicians tend to prescribe off-label when it is in the best interest of the
patient suggests that an out-right ban on off-label use, and possibly even the current ban on
promotion of off-label uses, has the potential to harm welfare. However, much more research
is needed for clear and effective policy to be developed.
For example, little is known about how the current FDA regulatory architecture affects
incentives for firms to invest in identifying new uses for existing drugs. The FDA provides
three-year exclusivities to incentivize firms to incur the costs associated with seeking new
indications for existing drugs, but this incentive may be negligible if the drug is already
generic. Currently, firms face a difficult decision of whether to rely on clinicians to discover
and promote these new uses, or incur the costs associated with adding the indication to a
drug’s label that will likely soon be, or already is, generic. Data from the FDA Orange Book
on new chemical entities, associated patents and associated exclusivities exist can shed light
on these questions. We look forward to further progress.
28
References
Adams, C.P.; Brantner, V. V. “Estimating the Cost of New Drug Development: Is It Really
$802 million?” Health Affairs 25, 2006, 420-428.
Aitken, M.; Berndt, E. R.; Cutler, D. M. “Prescription Drug Spending Trends in the United
States: Looking Beyond the Turning Point.” Health Affairs 28, 2009, 151-160.
Avenel, S.; Bomkratz, A.; Dassieu, G.; Janaud, JC.; Danan, C. “The Incidence of Prescriptions without Marketing Product License in a Neonatal Intensive Care Unit.”
Pediatrics, 7, 2000, 143-147.
Bradford, W.; Kleit, A.; Krousel-Wood, M.; Re, R. “Testing Efficacy with Detection Controlled Estimation: An Application to Telemedicine.” Health Economics, 10, 2001,
553-564.
Cuzzolin, L.; Zaccaron, A.; Fanos, V. “Unlicensed and off-label uses of drugs in paediatrics:
a review of the literature,” Fundam Clin Pharmacol 17, 2003, 125-31.
DiMasi, J, “Risks in new drug development: approval success rates for investigational
drugs,” Clinical Pharmacology and Therapeutics - St. Louis 69(5), 2001, 297-307.
DiMasi, J.; Hansen, R.; Grabowski, H. “The Price of Innovation: New Estimates of Drug
Development Costs,” Journal of Health Economics 22, 2003, 151-185.
Dranove, D. “Medicaid Drug Formulary Restrictions,” Journal of Law and Economics 32,
1989, 143-162.
Evans, D. “Pfizer Broke the Law by Promoting Drugs for Unapproved Uses,” Bloomberg
November 9, 2009.
Feinstein, J. “The Safety Regulation of U.S. Nuclear Power Plants: Violations, Inspections,
and Abnormal Occurrences,” Journal of Political Economy 97, 1989, 115-154.
Feinstein, J. “Detection Controlled Estimation,” Journal of Law and Economics 33, 1990,
233-276.
Feinstein, J. “An Econometric Analysis of Income Tax Evasion and Its Detection,” RAND
Journal of Economics 22, 1991, 14-35.
Harris, R. The Real Voice. New York: Macmillan, 1964.
Heidelbaugh, J.; Riley, M.; Habetier, J. “10 Billing & Coding Tips to Boost Your Reimbursement,” Journal of Family Practice 57, 2008, 724-730.
29
Helland, E. “The Enforcement of Pollution Control Laws: Inspections, Violations, and SelfReporting,” Review of Economics and Statistics 80, 1998, 141-153.
Huskamp, H.A. “Managing Psychotropic Drug Costs: Will Formularies Work?” Health
Affairs 22(5), 2003, 84-96.
Iizuka, T. “Prescribing Trends in Psychotropic Medications,” Journal of Industrial Economics 52, 2004, 349-379.
Klein, D.; Tabarrok, A. “Who Certifies Off-Label?” Regulation 2004, 60-63.
Kesselheim, A.; Mello, M.; Studdert, D. “Strategies and Practices in Off-Label Marketing
of Pharmaceuticals: A Retrospective Analysis of Whistleblower Complaints,” PLoS
Medicine 8(4), 2011, 1-9.
Kolata, G. “How Fen-Phen, A Diet ’Miracle,’ Rose and Fell,” The New York Times, September 23, 1997.
Kleit, A.; Ruiz, J. “False Positive Mammograms and Detection Controlled Estimation,”
Health Services Research 38, 2003, 1207-1228.
Lin, H.; Phan, K.; Lin, S. “Trends in Off-Label β-Blocker Use: A Secondary Data Analysis,”
Clinical Therapeutics 28, 2006, 1736-1746.
Mack, A. “Examination of the Evidence for Off-Label Use of Gabapentin,” J. Managed Care
Pharm. 9, 2003, 559-68.
McIntyre, J.; Conroy, S.; Avery, A.; Corns, H.; Choonara, I. “Unlicensed and off label
prescribing of drugs in general practice,” Archives of Disease in Childhood 83, 2000,
498-501.
Molitor, D. “Physician Behavior and Technology Diffusion in Health Care,” MIT Doctoral
Dissertation, 2012.
Oliver, T.R.; Lee, P.R.; Lipton, H.L. “A Political History of Medicare and Prescription Drug
Coverage,” The Milbank Quarterly 82, 2004, 283-354.
O’Reilly, J.; Dalal, A. “Off-Label or Out of Bounds? Prescriber and Marketer Liability for
Unapproved Uses of FDA-Approved Drugs,” Ann Health L 12, 2003, 295-300.
Peltzman, S. “An Evaluation of Consumer Protection Legislation: The 1962 Drug Amendments,” Journal of Political Economy 81, 1973, 1049-91.
Perkins, K.A.; Perkins, C.A.; Conklin, Levine, M.D. Cognitive Behavior Therapy for Smok-
30
ing Cessation: A Practical Guidebook to the Most Effective Treatments, Routledge:
New York, 2008.
Pincus, H.A.; Tanielian, T.L.; Marcus, S.C.; Olfson, M.; Thompson, J.; Zito, J.M. “Prescribing Trends in Psychotropic Medications,” Journal of the American Medical Association 279, 1998, 526-531.
Radley, D.; Finkelstein, S.; Stafford, R. “Off-label Prescriptions Among Office-Based Physicians,” Arch Internal Med 166, 2006, 1021-26.
Rowland, D. (Executive Director, Kaiser Commission on Medicaid and the Uninsured).
Prescription Drug Coverage for the Medicare Population - Testimony before the Subcommittee on Health, Committee on Energy and Commerce, The United States House
of Representatives. , February 15, 2001.
Salbu, S.R. “Off-Label Use, Prescription, and Marketing of FDA-Approved Drugs: an Assessment of Legislative and Regulatory Policy,” Fla. L. Rev. 51, 1999, 18.1
Stafford, R. “Regulating Off-Label Drug Use - Rethinking the Role of the FDA,” New
England Journal of Medicine 358, 2008, 1427-29.
Temin, P. “The Origin of Compulsory Drug Prescriptions,” Journal of Law and Economics
22, 1979, 91-105.
Thomas, C.P.; Conrad, P.; Casler, R.; Goodman, E. “Trends in the Use of Psychotropic
Medications Among Adolescents, 1994 to 2001,” Psychiatric Services 2006.
United States General Accounting Office. “Off-Label Drugs: Reimbursement Policies Constrain Physicians in Their Choice of Cancer Therapies,” September 27, 1991.
United States General Accounting Office. “Prescription Drugs: Implications of Drug labeling
and Off-Label Use,” September 12, 1996.
United States General Accounting Office. “Medicaid Outpatient Prescription Drugs,” December 15, 2010.
31
Table 1: Patient and Physician Variables
Patient
Patient
Patient
Patient
Patient
Patient
Patient
Patient
Patient
age
is female
is Hispanic
is African-American
is other (non-Caucasian) race
has Medicare
has Medicaid
has private insurance
has other insurance
Physician
Physician
Physician
Physician
Physician
Physician
Physician
specialty is internal medicine
specialty is pediatrics
specialty is OB/GYN
specialty is cardiology
specialty is psychiatry
specialty is neurology
has other specialty
Southern Census region
Mid-West Census region
West Census region
Observations
mean
sd
51.780
0.582
0.075
0.089
0.093
0.313
0.102
0.475
0.041
23.220
0.493
0.263
0.284
0.290
0.464
0.303
0.499
0.198
0.132
0.064
0.023
0.112
0.072
0.062
0.304
0.339
0.245
0.149
0.316
0.258
0.242
0.460
0.228
0.344
0.225
0.419
0.475
0.418
547,977
Note: These statistics reflect the demographic composition and insurance status of patients in our NAMCS
sample of prescriptions during 1993-2008, as well as the specialty and location of the physician treating them.
32
Table 2: Prescription and Indication Variables
mean
sd
0.335
71.490
3.872
0.617
0.335
0.887
0.485
0.472
69.290
2.244
0.486
0.472
0.316
0.500
Anti-infectives
Cardiovascular agents
Central nervous system agents
Coagulant modifiers
Gastrointestinal agents
Hormonal agents
Miscellaneous agents
Respiratory agents
Topical agents
Psychological agents
Immunologic agents
Metabolic agents
0.105
0.185
0.202
0.018
0.044
0.063
0.017
0.083
0.094
0.081
0.028
0.079
0.306
0.388
0.402
0.134
0.205
0.242
0.130
0.276
0.292
0.272
0.166
0.270
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
ICD9
0.035
0.039
0.154
0.130
0.137
0.254
0.177
0.058
0.066
0.074
0.123
0.137
0.046
0.185
0.192
0.361
0.337
0.344
0.435
0.382
0.234
0.247
0.261
0.329
0.344
0.209
Prescription observed to be on-label
Number of drugs approved to treat indications
Number of prescriptions written during visit
Two indications recorded on the NAMCS form
Three indications recorded on the NAMCS form
NAMCS form allows six prescriptions
NAMCS form allows eight prescriptions
code
code
code
code
code
code
code
code
code
code
code
code
code
for
for
for
for
for
for
for
for
for
for
for
for
for
infectious and parasitic disease
neoplasm
endocrine disorders
mental disorders
nervous system disorders
circulatory system disorders
respiratory system disorders
digestive system disorders
genitourinary system disorders
skin disorders
musculoskeletal system disorders
ill-defined disorders
injury and poisoning
Observations
547,977
Note: These statistics reflect information on the prescriptions and indications in our NAMCS sample
during 1993-2008. The top portion of the table provides information on the number of prescriptions and
indications, while the middle and bottom portions provide information on the types of drugs and indications,
respectively.
33
Table 3: Form Limitations and Detection, Mean Y2i
One Prescription
Two Prescriptions
Three Prescriptions
Four Prescriptions
Five Prescriptions
Six Prescriptions
Seven Prescriptions
Eight Prescriptions
Average
Observations
(1)
One Indications
(2)
Two Indications
(3)
Three Indications
(4)
Average
0.379
0.318
0.271
0.207
0.164
0.114
0.086
0.069
0.448
0.454
0.413
0.365
0.308
0.256
0.202
0.151
0.461
0.469
0.464
0.428
0.390
0.361
0.318
0.277
0.408
0.392
0.370
0.334
0.302
0.278
0.234
0.205
0.267
209,733
0.364
154,843
0.388
183,401
0.335
547,977
Note: These statistics reflect the mean of Y2i , the indicator for whether on-label use both occurred and
is detected, by number of prescriptions and indications recorded by the physician on the NAMCS survey
form.
34
Table 4: DCE Model for Detection of On-Label Prescribing
(1)
(2)
(3)
-0.120***
(-98.69)
0.011*
(1.71)
0.067***
(12.12)
-0.077***
(-9.59)
-0.069***
(-11.94)
-0.123***
(-99.13)
0.010*
(1.64)
0.067***
(12.03)
-0.073***
(12.03)
-0.061***
(-10.97)
-0.128***
(-99.86)
0.020***
(3.15)
0.045***
(8.01)
-0.067***
(-8.06)
-0.039***
(-6.72)
-0.003***
(-24.72)
-0.048***
(-9.53)
0.027***
(2.85)
0.130***
(15.20)
-0.050***
(-5.40)
-0.003***
(-24.62)
-0.048***
(-9.47)
0.026***
(2.72)
0.127***
(15.07)
-0.053***
(-5.75)
-0.003***
(-23.89)
-0.027***
(-5.20)
0.002
(0.24)
0.101***
(11.76)
-0.053***
(-5.62)
Indicators for Year
Quadratic Time Trend
Indicators for Physician-Specialty
Interacted Indicators for Year and Physician Specialty
On-Label
—
On-Label
—
—
On-Label
On-Label
—
—
—
Detection
On-Label
Observations
547,977
547,977
547,977
Exclusive to Detection Model
Number of prescriptions written during visit
Two indications recorded on the NAMCS form
Three indications recorded on the NAMCS form
NAMCS form allows six prescriptions
NAMCS form allows eight prescriptions
Common to Both Models
Patient age
Patient is female
Patient is Hispanic
Patient is African-American
Patient is other (non-Caucasian) race
Other Controls
Note: These coefficient estimates are for the model of detection. The top portion of the table reports
estimates for those variables that are exclusive to the model of detection, while the middle portion reports
those variables that are also in the model of physician on-label prescribing. The bottom portion of the table
indicates whether each of the various controls is included in the specification of the on-label and detection
models. Numbers in parentheses are t-statistics. *p < 0.10, **p < 0.05, ***p < 0.01.
35
Table 5: DCE Model for On-Label Prescribing
(1)
(2)
(3)
0.044***
(64.45)
-0.087***
(-5.06)
-0.140***
(-7.02)
-0.079***
(-5.69)
-0.087***
(-3.37)
0.043***
(67.24)
-0.088***
(-5.10)
-0.140***
(-7.06)
-0.084***
(-6.02)
-0.110***
(-4.08)
0.033***
(84.07)
-0.086***
(-5.05)
-0.126***
(-6.32)
-0.078***
(-5.61)
-0.070***
(-2.69)
-0.003***
(-9.07)
-0.076***
(-6.73)
-0.035*
(-1.72)
-0.110***
(-5.75)
-0.016
(-0.76)
-0.003***
(-9.43)
-0.072***
(-6.39)
-0.021
(-1.02)
-0.100***
(-5.28)
0.017
(0.79)
-0.002***
(-5.29)
-0.044***
(-4.01)
-0.000
(-0.01)
-0.079***
(-4.21)
0.026
(1.28)
On-Label
—
On-Label
—
547,977
—
On-Label
On-Label
—
547,977
—
—
Detection
On-Label
547,977
Exclusive to Model of On-Label Prescribing
Number of drugs approved to treat indications
Patient has Medicare
Patient has Medicaid
Patient has private insurance
Patient has other insurance
Common to Both Models
Patient age
Patient is female
Patient is Hispanic
Patient is African-American
Patient is other (non-Caucasian) race
Other Controls
Indicators for Year
Quadratic Time Trend
Indicators for Physician-Specialty
Interacted Indicators for Year and Physician Specialty
Observations
Note: These coefficient estimates are for the model of physician on-label prescribing. The top portion of the
table reports estimates for those variables that are exclusive to the model of physician behavior, while the
middle portion reports those variables that are also in the model of detection. Patients without insurance
are the excluded dummy variable. The bottom portion of the table indicates whether each of the various
controls is included in the specification of the on-label and detection models. Numbers in parentheses are
t-statistics. *p < 0.10, **p < 0.05, ***p < 0.01
36
Table 6: Form Limitations and Detection, Mean of Φ(x2i β̂2 )
One Prescription
Two Prescriptions
Three Prescriptions
Four Prescriptions
Five Prescriptions
Six Prescriptions
Seven Prescriptions
Eight Prescriptions
Average
Observations
(1)
One Indication
(2)
Two Indications
(3)
Three Indications
(4)
Average
0.652
0.602
0.549
0.495
0.440
0.379
0.317
0.269
0.649
0.602
0.549
0.495
0.442
0.383
0.322
0.273
0.663
0.618
0.568
0.514
0.464
0.408
0.344
0.297
0.653
0.605
0.555
0.502
0.451
0.395
0.332
0.286
0.539
209,733
0.509
154,843
0.465
183,401
0.506
547,977
Note: These estimates reflect the mean probability of detection, Φ(x2i β̂2 ), by number of prescriptions and
indications recorded by the physician on the NAMCS survey form.
37